P1-08-12: Hormonal Therapy Compliance and Mortality in Metastatic Breast Cancer.
Methods: A retrospective analysis was conducted using linked pharmacy claims and cancer registry data from the population-based National Cancer Registry Ireland (NCRI). Women with stage IV breast cancer at initial diagnosis (2004-2006), who received HTx as 1st line treatment for ≥90 days were identified (tamoxifen, toremifene, fulvestrant, anastrozole, letrozole, exemestane). Compliance was calculated over the time that patients remained on HTx i.e. from the date of treatment initiation to the earliest of either HTx discontinuation or end of follow up (31/12/07, 3 years or death). Patients discontinuing HTx were assigned their compliance rate at this time for the remainder of follow-up. In the primary analysis Cox proportional hazard models with late entry at 90 days post HTx initiation and time varying compliance (defined dichotomously as: low < 80%, high ≥ 80%), were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for all-cause and breast cancer specific mortality. Secondary analyses of (a) time to treatment discontinuation and (b) compliance in the first 90 days of treatment were conducted. All models were adjusted for age, tumor grade, number/site of metastases, estrogen, progesterone and human epidermal growth factor 2 receptor status. Results: 359 women with stage IV breast cancer at diagnosis were identified. Of these, 151 received HTx as 1st line treatment. Median HTx compliance was 84.8%, (range 19.4%-100%; interquartile range 75.5%-100%). Demographic, tumor and treatment characteristics were similar between low (n=51) and high (n=100) compliance groups. The number and distribution of metastatic sites at diagnosis was also similar (low/high compliance: bone only 70.6%/70.0%, lung only 17.6%/18.0%, liver only 11.8%/14.0%, brain only 0.0%/0.0%, other 13.7%/17.0%, multiple sites 15.7%/18.0%). Compliance rates for both compliance groups increased marginally over time. Unadjusted all-cause mortality rates were 143 and 359 deaths/1000 woman years for low and high compliance respectively. Low compliance was associated with a 90% increased risk of all-cause (HR 1.90; 95%CI 1.06, 3.41) and breast cancer specific (HR 1.91; 95%CI 1.03, 3.55) mortality. In the secondary analyses low compliance was associated with a shorter time to HTx discontinuation (HR 1.97; 95%CI 1.01, 3.84). Low compliance within the first 90 days of treatment was associated with non-significantly increased mortality (all-cause HR 1.57; 95%CI 0.88, 2.80; breast cancer specific HR 1.58; 95%CI 0.85, 2.93). Discussion: This study demonstrates, for the first time to our knowledge, a significant association between low HTx compliance and increased mortality in patients with MBC. In addition, the observed association between compliance and treatment duration is consistent with a reduced time to disease progression in low compliance patients. The findings do not appear to be related to either the burden of metastatic disease at diagnosis or to increasing disease severity causing lower compliance over time. These results suggest that the development and evaluation of effective interventions to support HTx compliance in patients with MBC is warranted. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-08-12.
Read more